# Trainer Api Training Arguments (info source where to look)

**URL:** <https://discuss.huggingface.co/t/trainer-api-training-arguments-info-source-where-to-look/89574>\
**Category:** Beginners\
**Created:** [May 31, 2024, 9:14am UTC](https://discuss.huggingface.co/t/trainer-api-training-arguments-info-source-where-to-look/89574 "2024-05-31T09:14:00Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![KaranHugger](https://avatars.discourse-cdn.com/v4/letter/k/db5fbb/32.png) [@KaranHugger](https://discuss.huggingface.co/u/KaranHugger)\
**Post date:** [May 31, 2024, 9:14am UTC](https://discuss.huggingface.co/t/trainer-api-training-arguments-info-source-where-to-look/89574/1 "2024-05-31T09:14:00Z")

</div>

Hello I am working with Trainer api and I wanted to know if you have a source where the Training arguments are better explained. I do not possess the vocabulary to understand what each of them mean. While training, I am facing a CUDA OOM error and I couldnt pinpoint which ones control the GPU memory usage.

---

<div class="post-metadata">

**Author:** ![nielsr](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/nielsr/32/39617_2.png) [@nielsr](https://discuss.huggingface.co/u/nielsr)\
**Post date:** [June 1, 2024, 10:32am UTC](https://discuss.huggingface.co/t/trainer-api-training-arguments-info-source-where-to-look/89574/2 "2024-06-01T10:32:44Z")

</div>

Hi,

Would recommend this guide for efficiently training on your GPU: [Methods and tools for efficient training on a single GPU](https://huggingface.co/docs/transformers/en/perf_train_gpu_one).

I’d recommend starting with a small batch size, use gradient accumulation/checkpointing, use a non-memory intensive optimizer like Adam-8bit, etc.

Also if you’re training an LLM it might help to use LoRa instead of full fine-tuning

---

<div class="post-metadata">

**Author:** ![KaranHugger](https://avatars.discourse-cdn.com/v4/letter/k/db5fbb/32.png) [@KaranHugger](https://discuss.huggingface.co/u/KaranHugger)\
**Post date:** [June 6, 2024, 1:20pm UTC](https://discuss.huggingface.co/t/trainer-api-training-arguments-info-source-where-to-look/89574/3 "2024-06-06T13:20:51Z")

</div>

I should point out the particulars I am using the pegasus-dailymail-cnn model. I have a small dataset of 14372, 819, 819 of train test val split. And currently my parameters are set to following and it is still causing me OOM errors. I will check out the given link. But I would really welcome any quick idea you can give me based on my current parameters  
 ![params](https://us1.discourse-cdn.com/hellohellohello/original/3X/d/7/d71c8d900cf73f93d9d96173940e4e0c7f51c48b.png)